In North America, the most common causes of a sepsis are bacteria such as Escherichia coli or Staphylococcus aureus. In addition to bacterial infection, fungal infections have in recent times become a significant cause of the disease. Fungal infections tend to be associated with higher rates of death. Only approximately 5% of fungal caused cases of sepsis are identified during the disease due to the poor diagnostic methods available. Recent studies have shown that patients with severe sepsis or septic shock showed an increased likelihood of death of 7.6% for every hour in which antibiotic therapy is not applied. Survival rates are also significantly reduced when antibiotics are not applied within the first 6 hours of identifying hypotension. Survival rates will be significantly increased if diagnosis times are reduced.
Culturing microorganisms from blood samples (Gram staining) remains the gold standard in detection of the microbiological cause of sepsis. This method is however subject to significant disadvantages, in particular, due to the large time difference between taking a blood sample and providing the results. It is not uncommon that 24 to 72 hours pass between taken a sample and providing diagnostic information. Within this time, broad band, untargeted antibiotic therapies are often introduced. This may lead to some success in treating the disease but is related to significant disadvantages with respect to the development of antibiotic resistant microorganisms.
Microarray and multiplex PCR approaches have been disclosed in the art, which are typically defined by extremely large numbers of probes or primers required for application of such methods (leading to significant cost and effort), a limited pool of target pathogens capable of being detected (such as only a limited sub-group of bacterial pathogens, or the absence of fungal pathogens), or a lack of discrimination between gram-negative and gram-positive bacterial pathogens, which provides sub-standard information for appropriate antibiotic therapies (US 2009286691 and US 201 1 151453). Methods for discriminating gram-positive and gram-negative bacteria have been disclosed in the art (US 20080118923 A1), in addition to the combined analysis of 16S and 18S sequences of bacteria and fungi (US 20090061446). Such methods, although potentially useful in clinical diagnostics, have never been applied in sepsis analytics and employ large numbers of primers in either microarray or very complex multiplex reactions, representing a significant technical and financial challenge for clinical diagnostic laboratories.
Multiplex RT-PCR approaches have been described in which a number of commonly occurring human pathogens are potentially detected. One example of such a multiplex PCR method is described in Gosiewski et al (BMC Microbiology 2014, 14:144) and U.S. Publication No. 2015/0232916, which discloses a nested PCR approach for detecting gram-positive and gram-negative bacteria, yeast, fungi and filamentous fungi from blood samples. A nested polymerase chain reaction involves two sets of primers, used in two successive runs of polymerase chain reaction, the second set intended to amplify a secondary target within the first run product. Nested PCR is applied in order to reduce nonspecific binding in products due to the amplification of unexpected primer binding sites, as it is unlikely that any of the unwanted PCR products contain binding sites for both the new primers in the second PCR run, ensuring the product from the second PCR has little contamination from unwanted products of primer dimers, hairpins, and alternative primer target sequences. Despite potentially reducing background signal, the PCR method described in Gosiewski and U.S. Publication No. 2015/0232916 are relatively complex and require two cycling reactions, essentially doubling the time, effort and reagents required for the analysis.
Other methods have employed the amplification of a number of PCR products from bacterial and fungal pathogens using sequence-specific oligonucleotides together with sequence-unspecific dyes, and subsequently, a melting curve analysis to differentiate between the various products (Horvath et al, BMC Microbiology 2013, 13:300). The method disclosed therein is however limited by a number of disadvantages known to occur with melting curve analyses.
Other methods have employed the amplification of a number of PCR products from bacterial and fungal pathogens using non-sequence-specific oligonucleotides together with sequence-specific probes to differentiate between the various products as described in EP 3172337. In such cases, only a broadband antibiotic therapeutic approach is possible, which may, in fact, be poorly suited for the particular pathogen.
Electrochemical detection techniques have higher detection sensitivity than conventional luminescence techniques (e.g., fluorescence and phosphorescence) due to higher signal-to-noise ratios. Because of their sensitivity and ability to accurately measures low-concentrations of nucleic acids, electrochemical detection techniques are able to differentiate between pathogenic species representing a significant technological improvement over the prior art. But, because of their sensitivity, false positive detection rates are high. Indeed, where organisms are cultured, the growth media often contains non-viable organisms or DNA/nucleic acids, which would not affect culture, but could produce false positives in PCR. If a system is designed uniformly for increased sensitivity to detect low titers pathogens, frequent false positive results may occur from background organisms or DNA/nucleic acids. Alternatively, if system sensitivity is reduced to avoid background organism detection, low titer organisms may be missed, resulting in false negative detection.
Further, when blood or other bodily fluids are obtained from a subject they may be contaminated by skin cells, bacteria, fungi, viruses, phages, their respective nucleic acids (including RNA and DNA) and/or other undesirable molecules, or disinfectants. Antiseptics are crucial for the practice of medicine; however, currently used antiseptics have a significant failure rate which results in substantial additional medical costs. Antiseptics are commonly used prior to routine phlebotomy, in preparation for minor and major invasive procedures, and as part of routine infection control hand-washing practices. The failure of antiseptics often result in erroneous diagnostic tests. For example, it has been estimated that a single false positive blood culture (i.e., where the culture indicates that the blood has been infected with bacteria, although the blood was contaminated during the blood draw) done on blood drawn from a patient at a hospital costs the patient an additional $2000 to $4,200 in unnecessary medication, additional follow up testing, and increased length of stay. (Bates, 1991).
Thus, there is a need in the art to provide methods which can selectively detect pathogenic organisms of interest. In particular, there is a need in the art for a method which enables the discrimination between a systemic infection and a false positive signal due to blood matrix bottle contamination. There is also a need in the art to identify when a blood culture is contaminated during blood draw.